# An Empirical Study on The Properties of Random Bases for Kernel Methods

###### tags: `papers`

MAIN IDEAS:
- Kernel machines and NN’s possess universal function approximation properties
- But ways of choosing the appropriate function class differ
- NN’s learn representation by adapting their basis functions to the data
- Kernel methods use a basis not adapted during training
- Contrast random features of approximated kernel machines with learned features of NN’s
- How do random and adaptive basis functions affect quality of learning?
- Present basis adaptation schemes that allow for more compact representation while retaining generalization properties of kernel machines